2931 papers
This paper proposes IBKA-Transformer, a novel fault diagnosis method that integrates an improved Black Kite Algorithm with a Transformer model to optimize hyperparameters and select features, achieving 99.86% accuracy in industrial robot bearing fault diagnosis.
The paper proposes DCUG-PPO, a dual-critic reinforcement learning controller that anchors its policy to a model-based fallback via an uncertainty-gated mechanism, achieving significantly higher reliability and smoother control than single-critic PPO and classical baselines in image-based visual servoing under camera dropout, while acknowledging limitations in hardware validation and broader algorithmic comparisons.
This paper introduces ESSIM, a novel probabilistic swarm optimization algorithm that leverages Bayesian statistical inference and individual particle uncertainty to maintain swarm diversity and avoid local minima, thereby outperforming classical PSO on multimodal benchmarks and Gaussian Process Regression hyperparameter optimization tasks despite requiring longer computation times.
This paper proposes a robust video steganography method that combines MEC-AQIM embedding in the DWT-SVD domain with a dual-layer pseudo-noise synchronization mechanism to effectively recover secret data from videos subjected to both recompression and arbitrary cropping.
This paper proposes a bio-inspired hybrid lifelong learning system that co-designs a Drosophila-human hybrid algorithm with self-selective memristor hardware to overcome catastrophic forgetting and energy inefficiencies, achieving significant speed and energy improvements over traditional CMOS accelerators.
This paper presents an auditable, leakage-controlled machine learning framework that harmonizes heterogeneous European emission data to accurately predict short-term spatial updates for cadmium, mercury, and lead inventories, achieving significant error reductions over baselines while explicitly defining operational limits for anomaly screening and expert prioritization rather than unrestricted forecasting.
This paper introduces bio2C, a FAIR-compliant benchmark of 2,500 curated natural language-Cypher pairs, and a modular Retrieval-Augmented Generation framework that significantly improves Text2Cypher accuracy for biomedical knowledge graphs by leveraging retrieved examples over schema-based prompting.
This study demonstrates that frontier large language models exhibit "performative self-correction" by generating extensive critical discourse without ever substantively revising their initial answers across 72 iterative review rounds, thereby challenging the efficacy of prompting-based correction strategies for improving AI reliability.
This paper presents a transparent, retrieval-assisted framework that grounds natural-language numeric evidence into structured optimization schemas, demonstrating that while schema retrieval is highly effective, the primary bottleneck for automating optimization problem instantiation lies in the downstream task of accurately matching extracted numbers to their corresponding slots.
This study introduces BoneNet, a multimodal deep learning framework that integrates a YOLOv8m backbone with dilated residual attention blocks and tabular data fusion to achieve high-accuracy automated detection, severity grading, and explainable depth estimation of peri-implant bone loss while transparently identifying performance limitations in severe cases.